Details

Autor(en) / Beteiligte

• Boegh, E
• Thorsen, M
• Butts, M.B
• Hansen, S
• Christiansen, J.S
• Abrahamsen, P
• Hasager, C.B
• Jensen, N.O
• van der Keur, P
• Refsgaard, J.C
• Schelde, K
• Soegaard, H
• Thomsen, A

Titel

Incorporating remote sensing data in physically based distributed agro-hydrological modelling

Ist Teil von

• Journal of hydrology (Amsterdam), 2004-02, Vol.287 (1), p.279-299

Ort / Verlag

Amsterdam: Elsevier B.V

Erscheinungsjahr

2004

Link zum Volltext

Quelle

ScienceDirect

Beschreibungen/Notizen

• Distributed information on land use and vegetation parameters is important for the correct predictions of evapotranspiration rate and soil water balance while, in turn, the growth and function of vegetation are also highly dependent on the soil water availability. In this study, the relationship between the soil water balance and the vegetation growth is represented by coupling a hydrological model (MIKE SHE) and a vegetation-SVAT model (Daisy) which simulates the interactions between soil, vegetation and atmosphere including the seasonal variation in plant structure and function. Because the coupling of process models is accompanied by increasing difficulties in obtaining values for the numerous parameters required, the utility of satellite data to set up, verify and update such a model system is the focus of the present paper. To achieve spatially distributed information on surface conditions, field data of leaf area index ( L) and eddy covariance fluxes were collected, and high-resolution remote sensing (RS) data were acquired to produce maps of land cover, leaf area index and evapotranspiration rates ( E). The land cover map is used to set up the model which is run throughout 1998 for a Danish agricultural area with a time step of 1 h. In May, the spatial heterogeneity of the leaf area index is at its largest, and the model performance is evaluated in time and space using the field measurements and the RS-based maps of L and E. Finally, the effect of adjusting the simulated L to match the RS-based L is investigated. The adjustment strategy includes synchronization of all vegetation parameters to maintain congruity of the model canopy representation. While the predicted crop yields were improved, a large micro-scale spatial heterogeneity in L within the operational modelling units restricted improvements in the simulated E. The delineation of modelling units that are homogeneous with respect to the assimilated variable, L, requires separation of land use classes with respect to the temporal development in vegetation cover.